Samples of 6 whole-grape juices produced in a few regions in Brazil, marked with the acronym of the state where they were produced, have been purchased. The acronyms are the following: RS (2 samples: RS₍₁₎ and RS₍₂₎), RJ, GO, PR and SP. The eligibility criteria were: i) whole red grape juice; (ii) no added preservatives, stabilizers or antioxidants; (iii) no added water; (iv) no added sugar; and v) samples whose ingredients were only “100% grape juice”. The juice samples were selected at random from the main producer states. We did not have access to any samples produced in the North and Northeast regions.

For the spectrophotometric and electroanalytical experiments, each sample was diluted in the concentration of 10% in an ethanolic solution. All tests were performed in triplicate.

The radicals ABTS.+ (2,2 azino-bis-(3-ethylbenzothiazoline-6-sulfonate), DPPH (2,2-diphenyl-1-picrylhydrazyl) and the reagents used in the analyzes were all analytically pure and sourced from Sigma Chemical Co. (St. Louis, MO, USA). All electrolytic solutions were prepared with purified water using the Millipore Milli-Q system, conductivity ≤ 0.1µS.cm^-1 (Millipore S.A., Molsheim, France). Other reagents were purchased off-the-shelf.

The purple radical DPPH (2,2-diphenyl-1-picrylhydrazyl), when reduced by the antioxidant, is discolored and turns yellow. Reduction of radical DPPH is followed by a decrease in absorbance by 517 nm. The amount of analyte in µL required to reduce the DPPH absorbance by 50% (Efficient Concentration, EC50) is calculated to evaluate the antioxidant capacity of each sample and the free radical sequestering activity or percentage of discoloration. The analytes were solutions of 10% whole juices and the standard solution of gallic acid evaluated at different concentrations. Absorbance was monitored with a UV-Vis spectrophotometer (Jasco^® V-530) and all tests were performed in triplicate.

The radical ABTS.+ (2,2' azino-bis-(3-ethylbenzthiazoline-6-sulfonate) is formed from the ABTS reaction (7 mM) with 88 µl of potassium persulfate (140 mM) in the absence of light. To perform the analyses, a volume of 300 µL of the ethanolic analyte solution was placed in a test tube with 2.7 mL of the ABTS radical, then the tubes were covered with Parafilm^® and kept in the dark for 20 minutes and the absorbance was monitored at 734 nm using a UV-Vis spectrophotometer (Jasco^® V-530) and all tests were performed in triplicate.

The method Folin-Ciocalteau (FC) allows the quantification of phenolic compounds by evaluating the reduction of the FC reagent by these antioxidants with the formation of a blue complex whose intensity increases linearly at 765 nm. The total amount of phenols was obtained by constructing a standard curve prepared with gallic acid. The total phenolic content was expressed as µg equivalents of gallic acid (GAE) per ml of sample. All analyses were done in triplicate.

To perform the electrochemical measurements, an µAutolab III potentiostat/galvanostat integrated with the software GPES 4.9 (Eco-Chemie, Utrecht, The Netherlands) was used. The analyses were carried out in an electrochemical cell with capacity for 3 mL of a solution operated with a three-electrode system, an Ag/AgCl reference electrode (KCl 3 mol.L^-1), an auxiliary platinum electrode and, as a carbon paste electrode prepared with 0.075 g of carbon and 0.035 g of mineral oil was used as a working electrode.

The tests using differential pulse voltammetry (DPV) were performed with pulse amplitude of 50 mV and scanning velocity of ʋ = 10mVs^-1. The analyses were carried out in an electrochemical cell containing 1.75 ml of phosphate buffer pH 7.0 (0.1 M), adding 25 µL of the ethanolic solution of the analyte. The electrochemical index (EI) of the samples was calculated by summing up the result of each division of the current by the potential (I/E) for each anodic peak observed on the differential pulse voltammograms.

Male Wistar rats (200-230 g) from the UFG central vivarium were used. All experimental protocols respected the protocols approved by the UFG Research Ethics Committee (protocol: 044/17). This study is in line with the European Union Guide for the Care and Use of Experimental Animals (2010/63/EU).

The rats were euthanized by exsanguination under inhalational anesthesia (n = 5-6). The thoracic aorta was isolated, separated from the connective and adipose tissues and cut into rings (± 4 mm), which were mounted between two metal hooks, one of which was connected to a power transducer to record the isometric voltage (DATAQ Instruments, Akron, OH, USA) and the other was attached to the vial for the isolated organ containing modified Krebs solution [composition in mM: NaCl, 130.0; KCl, 4.7; KH 2 PO 4, 1.2; CaCl₂, 1.6; MgSO₄, 1.2; NaHCO₃, 14.9; glucose, 5.5], pH 7.4, under gasification with carbon dioxide (95% O₂ + 5% of CO₂) at 37°C and maintained at baseline at 1.5 g (optimal rest tension, as previously standardized in our laboratory).

After 60 minutes of basal tension stabilization, the arteries were pre-contracted with phenylephrine (0.1 µM) and the presence of endothelium was determined using ACh (1 µM). The rings were disposed of when the relaxation for ACh was smaller than 80%. The whole grape juices analyzed were filtered on a filter paper under light and immediately divided and frozen at -20°C for future experiments.

The arteries with intact precontracted endothelium (phenylephrine, 0.1 µM) were stimulated to relax by increasing the concentration of juice directly in the bath solution (0 to 30 µL/mL). As wine is popularly known throughout the world as a vasodilator, we also perform relaxation curves stimulated with Cabernet Sauvignon red wine, produced in France in 2015, in order to compare the vascular effect.

In another experiment series, the vasodilatory effect of the juice samples was repeated after treatment with the NOS, L-NAME (100 µM, 30 min) inhibitor in order to determine the participation of NO in the mediated juice-stimulated vasodilation.

The charts were made and analyzed by the software GraphPad Prism (GraphPad Software Corporation, version 5.0) by ANOVA plus Bonferroni post-test. In the analyses, the continuous variables showed normal distribution and the results were expressed as the mean ± standard deviation of the mean of at least five experiments (n = 5-6) obtained from different animals. All analyses considered a statistical significance level of 5% (p < 0.05).

Antioxidant activity was evaluated using two different methods: DPPH and ABTS. Although the sample from RJ had the best result (lower CE₅₀), we can see in Figure 1A that the four samples obtained from RS₍₁₎, RS₍₂₎, RJ and GO had similar DPPH discoloration indexes (CE₅₀: 18.2 ± 0.8; 22.4 ± 2.1; 17.1 ± 0.6; 20.5 ± 0.7 µL, respectively), indicating equivalent antioxidant activity. The samples from PR and SP presented similar results comparing each other, and lower antioxidant capacity (p < 0.05) compared to the other samples (CE₅₀: 29.1± 1.1; 28.6 ± 1.6, respectively).

Figure 1

The ABTS technique (Figure 1B) confirmed the best antioxidant activity of the RJ sample (CE₅₀: 0.59 ± 0.11 µL). The SP sample had a lower antioxidant potential (CE₅₀: 1.25 ± 0.19 µL), being statistically different (p < 0.05) from the best sample (RJ). In turn, the other samples presented similar results (p > 0.05). The order of activity for the ABTS TEST was RJ > RS₍₁₎ (0.83 ± 0.1 µL) > RS₍₂₎ (0.91 ± 0.18 µL) > GO (0.65 ± 0.01 µL) > PR (0.90 ± 0.13 µL) > SP.

In Figure 1C, we can see the number of phenolic compounds (expressed in µg equivalents of gallic acid/mL of the sample) present in the samples. The results are according to the antioxidant activity (Figures 1A, B), and the sample with the highest antioxidant potential presented a higher concentration of phenolic compounds (RJ: 2.25 ± 0.06 µg/mL). The lowest concentration of phenolic compounds was observed in the SP sample (1.26 ± 0.08 µg/mL), being different (p < 0.05) from all others.

Figure 2A shows the results of the electroanalytical analyses of the samples. Although the electrochemical indices showed a higher activity for the RJ sample (21.7 ± 2.1 µA/V), the samples from RS₍₁₎, RS₍₂₎, GO and PR presented equivalent electrochemical indices (p > 0.05) (18.8 ± 1.9; 16.3 ± 1.5; 19.8 ± 2.5; 17.2 ± 3.3 µA/V, respectively). Only the SP sample (11.3 ± 0.5 µA/V) showed a statistical difference (p < 0.05) compared to most effective sample (RJ).

Figure 2

The IE values ​​are calculated from the main electrochemical parameters, that is, an anode peak potential - E_pa. The lower the anode peak potential the higher the thermodynamic capacity of the species to reduce others, and anodic peak current intensity - I_pa - which in addition to being related to concentration, also concerns the speed with which the reduction process occurs. The presence of electroactive compounds in both the most effective sample (RJ) and in the least effective sample (SP), which presented two oxidation peaks (1_a and 2_a). The oxidation peak of the RJ sample occurs at a lower voltage than the SP sample, showing its greater reducing potential. (Figure 2B).

Vasodilator activity was demonstrated in vitro for the 6 juice samples tested and the red wine sample (for comparison purposes). All 6 juice samples induced concentration-dependent relaxation, whose efficacy was variable among them (Figure 3A). The RJ sample (as well as in the antioxidant activity tests) was the most effective one, inducing maximum relaxation of 88.2 ± 4.8%. This sample was the only one that had a similar maximum relaxation (p > 0.05) compared to red wine induced relaxation (112.9 ± 5.1%). The other samples (except RJ) showed lower relaxation (p < 0.05) than that produced by red wine. The RS₍₂₎ sample was the least effective one in inducing vascular dilatation (19.5 ± 6.1%), being significantly (p < 0.05) different from the RS₍₁₎ (71.6 ± 9.6%), RJ (88.2 ± 4.8%) and GO (53.6 ± 7.2%) samples. The PR and SP samples induced relaxation in similar magnitudes (39.8 ± 7.4 and 39.9 ± 10.4%, respectively) and only showed a significant difference (p < 0.05) over the RJ sample. These results are presented in figure 3B.

Figure 3

Regarding the mechanism of action, all juice samples tested revealed a major participation in the endothelial NO in the induction of relaxation, as L-NAME nearly completely inhibited relaxation (Figure 4). Only the RS₍₂₎ sample showed no difference (p > 0.05) after inhibition of NO with L-NAME, revealing an action mechanism unrelated to NO release by the vascular endothelium.

Figure 4